453
Organic and Biological Chemistry Organometallic Electrochemistry. I. Derivatives of Group IV-B Elements R. E. Dessy, William Kitching, and Tristram Chivers Contribution f r o m the Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221. ReceivedJuly 19, 1965 Abstract: Organometallic derivatives of group IV elements have been studied via polarography, controlled
potential electrolysis, and triangular voltammetry. Some subvalent organometals and anions have been generated and examined. In particular, the triphenylmetal radicals, PhaM. (where M = Si, Ge, Sn, Pb), have been shown to suffer quite dissimilar fates in dimethoxyethane. Abstraction (Si, Ge), coupling (Sn), and arylation of the Hg electrode (Pb) have been observed. Organometallic assemblies of the type Ph3M-Bu4N+(where M = Ge, Sn) have been shown to serve as electron-transfer reagents in reactions with other group IV halides rather than as nucleophiles. Reduction of RzSnClz proceeds via (R2SnCI),which rapidly dimerizes, providing a route to R2Sn(Cl)Sn(CI)RZspecies. Reduction of Ph2MXzleads to Ph2MHz(M = Si, Ge), or arylation (Pb).
A
lthough the polarographic behavior of a large number of organic molecules has been investigated successfully, and the electrochemical behavior made use of in determining energy levels, or applied synthetically, no extensive study and only one intensive exploration of the electrochemistry of organometallic species have been made (vide infra). Conceptually, one can envisage, for the reductive mode only, the addition of an electron or electrons to an organometallic assembly RmQ, where R is a u- or 7r-bonded organic moiety, m is a metal and two of its valency positions, and Q is another R group, halo or oxyfunction. The fate of the affected assembly may be to exist for long time periods as the radical anion, or to lose the ligand Q as an anion, generating the subvalent organometal Rm. The fate of Rm might then take one of several courses: (1) ultimate decomposition to m, R . and m ; ( 2 ) disproportionation, RmR which may be concomitant with 1 ; (3) abstraction of H from the surroundings, yielding RmH; or (4) coupling to yield RmmR. Further reduction may yield Rm, which could decompose. We have made an intensive study of the electrochemistry of organometallic compounds derived from the metals of group 11, 111, IV, V, and the transition series, in dimethoxyethane, using tetrabutylammonium perchlorate as a supporting electrolyte. High resistances in the H cell employed, where cathode and anode were separated by a glass frit, and low conductivity of the solution necessitated three electrode geometry. A 0.001 M AgC104 (0.1 M Bu4NC104)IAg reference electrode was employed, along with a mercury pool anode, and either a dme, sheared dme, hanging Hg drop, or Hg pool cathode. Polarographic data were obtained with a Metrohm Polarecord, or with a unit based on standard operational amplifier circuitry. The latter unit was also equipped to apply slow-single or multiple-triangular voltage sweeps to a hanging drop or higher frequency single or multiple-triangular voltage
+
(1) P. Delahay, “New Instrumental Methods in Electrochemistry,” Interscience Publishers, Inc., New York, N. Y., 1954.
sweeps to a sheared dme, during the last 100 to 1000 msec. of the drop’s life. This latter technique allows the detection of short-lived species at the electrode by subsequent reoxidation to the normal valence state. In the present system charging current problems set the lower limit on predecomposition lifetimes at 10-50 msec. The often high degree of electrochemical irreversibility (to be differentiated from chemical reversibility) led to the evaluation of n (number of electrons utilized at each polarographic step) via exhaustive controlled potential electrolysis. Where long-lived species seemed to be involved, such reduced solutions were electrochemically oxidized, often successfully, to regenerate the starting compound. Where stable intermediates or subvalent species were found, the chemistry of these species was explored, the course of the reaction being followed by polarography and ultraviolet spectroscopy. The present paper reports on systems derived from group IV metals. Experimental Section Equipment. Polarographic and controlled potential electrolysis experiments were made in an H cell, the anodic and cathodic compartments being separated by a 25-mm. (medium) glass frit. The arms of the cell were as close as possible, and had flat bottoms, measuring 3 cm. in diameter. Ground-glass outer joints and 0ringed Teflon adapters permitted the insertion of degassing tubes, dropping mercury electrode (dme), or hanging drop accessories. The reference electrode, a 0.001 M AgC104 (0.1 M Bu4NC10a) I Ag wire, made electrical contact with the cathode environment via a tube separated into compartments by two glass frits, the terminal one being as close (1-2 mm.) as geometrical factors permitted. Normal cell resistances were RT - R = 10 kohms, RT - c = 10 kohms. Electrical contact with the anode pool, and the cathodic pool, if used, was made via Pt foil pinch sealed through the H cell arms, at the base of the cell. The dme was a rather fast, 3-sec. capillary, longer drop times leading to excessive maxima, and gross irregularities in i/E curves. The hanging drop assembly was a microsyringe mounted on a standard capillary. Routine polarographs were made using a Metrohm Polarecord, in conjunction with a Sargent IR Compensator. Alternatively a (2) Cy. J. Rika in “Progress in Polarography,” P. Zuman and I. M. Kolthoff, Ed., Interscience Publishers, Inc., New York, N. Y.,1962.
Dessy, Kitching, Chivers
Derivatives of Group IV-B Elements
454 unit based on standard operational amplifier circuitry3 could be employed, feeding into a Mosely 2D-2 X-Y plotter. The unit possessed an auxiliary input capable of accepting externally generated wave functions which then could be applied to the cell. Triangular voltammetry therefore was performed by external generation of a variable frequency square wave, via alternate saturation of two switching transistors. Integration of the square wave led to the triangular signal desired, which also was used as a feedback to the square wave generator section. Appropriate controls in this section permitted adjustment independently of cathodic and anodic limits of the triangular wave, and its midpotential. The anodic and cathodic duration of the square wave also could be altered, thus permitting generation of asymmetric triangular waves.3 Connection directly to the bases of the switching transistors permitted gating of the generator by a monostable multivibrator.4 The latter could be actuated manually, or by a suitable signal from an astable rnultivibrator, which also synchronized the gate with actuation of the shearing solenoid attached to the dme.4 The former was used for hanging drop voltammetry, the latter for sheared drop. The readout devices were a Mosely 2D-2 recorder and a Tektronix Storage oscilloscope equipped with Polaroid camera. Controlled potential electrolysis was achieved at low current levels (10 ma.) via circuits based on conventional operational amplifier design. Higher levels (1000 ma.) were achieved with suitable master/slaving of Kepco CK-60-0.5 power supplies, the master being boot strapped to reduce drain on the reference electrode.6 Chemicals. Dimethoxyethane (Ansul 121) was purified by distillation through a 100-cm. packed column from LiAlH, under a nitrogen atmosphere. Careful attention to detail is necessary to prevent peroxide or water contamination. Tetrabutylammonium perchlorate was prepared from the metathetical reaction between tetrabutylammonium iodide and silver perchlorate in hot ethanol. The silver iodide was filtered o f f , the solution cooled to -80", or precipitated with water, and the perchlorate salt was filtered o f f and washed with water until freed from Ag+ or I-. Recrystallization from dimethoxyethane gave suitable material, with a window from approximately +0.9 to -3.6 with respect to reference. The material was dried in an Abderhalden pistol over P 2 0 5 ,using boiling xylene, for a period of several hours at 0.1 mm. Samples from G. F. Smith and Matheson Coleman and Bell also proved satisfactory. The organosilicon derivatives were kindly supplied by Dow Corning Corp. and were purified by standard methods of distillation or recrystallization. Organogermanium, -tin, and -lead compounds were largely supplied from the Organic Chemical Institute, Utrecht, Holland, and M and T Chemical Co., Rahway, N. J. Professor Max Schmidt of the Institute for Inorganic Chemistry of the University of Marburg very kindly supplied the compounds of the type (CsH5)&iM'M"(CSH5)3 where M = Sn; M' = Sn, Te; and M" = Sn, Ge.
Results and Discussion Recent years have witnessed a renaissance of interest in the organic chemistry of group IV elements both on t h e ~ r e t i c a l ~and - ~ syntheticlo,l 1 grounds. Because of current interest in the area, certain group IV-B derived organometals were chosen first for electrochemical study. Previous polarographic studies in this area have been few in number and restricted almost entirely to organotin compounds, The first recorded polarographic reduction of an organotin compound was that of diethyl(3) C. Enke, Princeton University, personal communication. (4) Electronics Design Shop, University of California, Los Angeles, California, personal communication. ( 5 ) P. Birman, Kepco Inc., personal communication. (6) A. L. Allred and E. G. Rochow, J . Inorg. Nucl. Chem., 5 , 269 (1958). (7) R . Drago, ibid., 15, 237 (1960). (8) A. L. Allred and E. G. Rochow, ibid., 20, 167 (1961). (9) For a theoretical discussion see D . P. Craig and C. Zaul, J . Chem. Phqs., 37, 609 (1962). (10) See H. Gilman, 0. L. Marrs, and S. Y. Sim, J . Org. Chem., 27, 4232 (1962), for discussion and leading references. (11) C . C. Tamborski, F. E. Ford, and E . J . Soloski, ibid., 28, 181 ( 1963).
Journal of the American Chemical Society I 88:3
tin dichloride by Riccoboni and Peboff,12who claimed to have observed one two-electron reduction wave. The postulated product of this reduction was diethyltin, which was assumed to possess quite high stability under the conditions. Costal3 studied a range of trialkyltin halides in 3 0 x isopropyl alcohol and obtained three reduction waves. The first was thought to be the formation of hexaalkylditin. Doubt seemed to surround the origin of the latter two waves. Toropova and Saikina, l 4 studied similar compounds in 40 aqueous ethyl alcohol solutions at varying pH. The ease of reduction as a function of the organic radicals was in the order ethyl > propyl > butyl. The discharge of the methyl derivatives was sensitive to conditions, a not surprising observation considering the hydrolytic capabilities of the system. Employing oscillographic polarographic techniques, Saikinals obtained data on the reversibility of the above reductions, and found them to be highly irreversible. Conceptually at least, the most thorough study of the polarographic behavior of organotin compounds must be credited to Allan l 6 who, utilizing dimethylformamide as solvent and tetrabutylammonium perchlorate as supporting electrolyte, examined a variety of structural types. A plague of practical problems, however, arising from the presence of traces of water in the solvent, allowed none but most general of conclusions. n values were estimated from id values, using approximate diffusion coefficients. Allan could place confidence in data for only three compounds, tributyl-, triphenyl-, and trimethyltin halides. For compounds of the type R3SnX and R2SnX2,the first reduction wave was postulated to involve one electron as shown.
+ e +R3Sn. + Cl-
R3SnCI
2R3Sn. +R3SnSnR3
+ e +R2Sn. + C1-
R2SnCI2
2R2Sn. +R2Sn-SnR2
I
c1
I 1
c1 c1
The second reduction waves were postulated to involve two electrons, the electrode reactions being R3SnCl
+ 2e +R2Sn + R:- + C1-
and
+ 2e +R2Sn + 2Cl-
R2SnCI2
The present work will support some of this speculation. R i c ~ o b o n ihas ~ ~ reported on the polarographic behavior of triethyllead chloride in aqueous potassium chloride media. A one-electron reduction was claimed leading to triethyllead. More recently, Curtis and Allred18 have obtained data on compounds of the type ~ - ( C H & M C ~ H ~ C ~ H E (12) L. Riccoboni and P. Peboff, A f r i I s t . Venefo Sci., 107, 123 (1949); Chem. Abstr., 44, 6752a (1950). (13) G. Costa, Gazz. Chirn. Ital., 80, 42 (1950); Chem. Abstr., 44, 9926d(1950). (14) V. F. Toropova and M. I
